Download Atmospheric CO2 targets for ocean acidification perturbation

Part 2: Experimental design of perturbation experiments
3
Atmospheric CO2 targets for ocean acidification perturbation experiments
James P. Barry1, Toby Tyrrell2, Lina Hansson3,4, Gian-Kasper Plattner5 and Jean-Pierre Gattuso3, 4
Monterey Bay Aquarium Research Institute, USA
National Oceanography Centre, University of Southampton, UK
3
Laboratoire d’Océanographie, CNRS, France
4
Observatoire Océanologique, Université Pierre et Marie Curie-Paris 6, France
5
Climate and Environmental Physics, University of Bern, Switzerland
1
2
3.1
Introduction
Research on ocean acidification has a primary goal of advancing our understanding of the consequences
for marine organisms and ecosystems of future changes in ocean chemistry caused by the anthropogenic
rise in atmospheric carbon dioxide levels. Though interesting as a basic research theme, ocean acidification
science should play a key role in the development of national and international policies for reducing CO2
emissions. To communicate the science of ocean acidification effectively to industrial leaders, the public
and policy makers, the science community must present the results and research implications in clear
and consistent terms that relate directly, if possible, to terms used currently in climate discussions, such
as atmospheric CO2 levels and potential stabilisation targets. To this end, ocean acidification research
programmes should be considered, designed, and reported in the context of realistic ranges for atmospheric
p(CO2) levels.
Throughout the brief history of ocean acidification research, a spectrum of p(CO2) levels has been used
for CO2 perturbation experiments. Though most studies involve the classic levels of ambient (~380 ppm)
and future atmospheric p(CO2) (usually 750 ppm, corresponding to the p(CO2) level expected by 2100),
values as low as 45 ppm (Buitenhuis et al., 1999) and as high as 150,000 ppm (Kikkawa et al., 2003)
have been reported. As long as the scientific community continues to work at different levels of p(CO2),
it will be difficult to integrate this information and develop coherent recommendations for policymakers
concerning the impacts of fossil fuel emissions on the oceans. To promote direct comparability among
studies and provide a clear link to atmospheric targets relevant to climate policy development, it is
advantageous to select key values of atmospheric p(CO2) for use as primary targets in ocean acidification
experiments. The range of these key values should overlap and span the range of present and future
atmospheric levels. Although inclusion of a much broader range of atmospheric p(CO2) levels may be
required for various reasons, the greatest societal benefit from ocean acidification studies is likely to lie
in increasing our understanding of the functional responses of species, populations, and ecosystems to
changes in ocean chemistry that could possibly occur over the next centuries.
Unlike atmospheric p(CO2), which is relatively homogeneous over the Earth, aqueous p(CO2) and other ocean
carbonate system parameters can vary greatly over space and time (Figure 3.1; Table 3.1). Water temperature
and salinity influence the solubility of carbon dioxide in seawater, widening the range of variability in the
carbonate chemistry of the oceans, particularly with latitude (Figure 3.2). In deep waters, the accumulation
of respiratory carbon dioxide increases the pools of total dissolved inorganic carbon (DIC1) and p(CO2),
leading to lower pH, lower carbonate ion (CO 2−
3 ) levels, and reduced saturation states for calcite (Ωc) and
aragonite (Ωa). Consequently, pH typically decreases with depth over much of the world ocean, particularly in
oxygen minimum zones. In the Eastern Pacific, low-pH waters from oxygen minimum zones upwell over the
continental shelf (Feely et al., 2008).
€
1
Chapter 1 provides detailed information on the chemical terms and symbols used in the present chapter.
Guide to best practices for ocean acidification research and data reporting
Edited by U. Riebesell, V. J. Fabry, L. Hansson and J.-P. Gattuso. 2010, Luxembourg: Publications Office of the European Union.
53
Part 2: Experimental design of perturbation experiments
Figure 3.1 Carbon dioxide concentrations over the ocean. A. Atmospheric p(CO2) levels (ppm). B. Surface
p(CO2) (µatm). Note the change in scale among plots. Data from Takahashi et al. (2009).
Temporal variation in carbonate system parameters can also be large, over scales as short as diel cycles (Bensoussan
& Gattuso, 2007; Wootton et al., 2008), over seasons (Kleypas et al., 2006; Findlay et al., 2008), and even in
association with episodic ENSO events (Friederich et al., 2002). As the balance between photosynthesis and
respiration varies over diurnal, seasonal, or other periods, there is a corresponding variation in DIC, pH, and other
ocean carbonate system parameters. DIC removal into phytoplankton blooms leads to an elevation of pH, [CO 2−
3 ],
Ωc and Ωa during late spring and summer. This is seen in model results and also in data from the Norwegian Sea and
in the seas to the north and west of Iceland (Findlay et al., 2008). Distinct, regular and repeated seasonal cycles are
also seen at sites closer to the equator, for instance at BATS and HOT (Kleypas et al., 2006). The use of a standard
€
set of p(CO2)atm targets, converted to p(CO2)aq or other carbonate system parameters in the habitat of concern (e.g.
ΩA in tropical surface waters or pH of temperate abyssal depths) will allow reporting of the relevant in situ carbonate
system parameters while maintaining a link to the common currency and units of climate change policy.
54
Mean
(present-day)
Observed range
(present-day)
Units
[DIC]
2020
1840 to 2200
µmol kg-1
strong variation with latitude
[AT]
2310
2170 to 2460
µmol kg-1
moderate spatial variability
[CO 2−
3 ]
203
81 to 302
µmol kg-1
strong variation with latitude
Ωc
4.9
2.0 to 9.1
strong variation with latitude
Ωa
3.2
1.2 to 5.4
strong variation with latitude
pH
8.1
7.9 to 8.7
strong spatial variability
p(CO2)
333
127 to 524
[Ca2+]
Parameter
€
Notes
µatm
strong spatial variation
10600
µmol kg-1
little spatial or seasonal
variation
[Mg2+]
55000
µmol kg-1
little spatial or seasonal
variation
Temp.
18.6
-1.9 to 29.6
°C
Salinity
34.8
10.8 to 37.5
-
[PO 3-4 ]
0.53
0.01 to 2.11
µmol kg-1
strong variation with latitude
[SiO2]
7.35
0.37 to 101
µmol kg-1
high in Southern Ocean
strong variation with latitude
moderate spatial variation
€
Table 3.1 Mean and range of variation in the main ocean carbonate system parameters over open ocean
surface waters of the world. This table can be used as a reference in the design of experiments including near
present-day (1990’s) carbonate system values (future conditions will obviously differ for some parameters).
The ranges of total dissolved inorganic carbon (DIC) and AT (total alkalinity) are from the GLODAP database
(Key et al., 2004) and represent surface water (0 m) of the open ocean, i.e. excluding coastal, shelf, and
enclosed seas, near-shore, and estuarine environments. Other carbonate system parameters were calculated
from DIC and AT using the csys software (see main text); temperature (Stephens et al., 2001) and salinity
(Boyer et al., 2001) were taken from the World Ocean Atlas (Antonov et al., 2006; Garcia et al., 2006;
Locarnini et al., 2006) database.
The goal of this chapter is to provide an overview of factors that influence the choice of atmospheric CO2 levels
used in ocean acidification studies, based on experimental design, target environments, location, and analytical
approach. The overriding philosophy for these guidelines is that ocean acidification research should attempt to
provide predictive capabilities concerning the response of the oceans, including its physics, biochemistry, and
biology, to a realistic range of future atmospheric p(CO2) levels.
3.2
Approaches and methodologies
We investigate the issue of target levels of atmospheric p(CO2), discuss the conversion of atmospheric values
into equivalent parameters of ocean carbonate chemistry, and then provide recommendations depending on the
number of treatment levels that can be manipulated.
55
Part 2: Experimental design of perturbation experiments
Figure 3.2 Distribution of ocean carbonate system parameters over the world ocean during the 1990’s. Total
dissolved inorganic carbon (DIC) and AT (total alkalinity) are from the GLODAP database (Key et al., 2004), and
represent surface water (0 m) of the open ocean. Other carbonate system parameters were calculated from DIC and
AT using the csys software (see main text) and temperature (Stephens et al., 2001) and salinity (Boyer et al., 2001)
from the World Ocean Atlas database. Plots courtesy of Andrew Yool.
3.2.1
Selection of key p(CO2)atm values
Key values for p(CO2)atm used in ocean acidification experiments should be based mainly on reasonable future
trajectories of atmospheric p(CO2), including intermediate stabilisation targets. The Intergovernmental Panel
on Climate Change (IPCC) in the Special Report on Emissions Scenarios (SRES; Nakićenović & Swart, 2000)
56
Figure 3.3 Atmospheric p(CO2) levels from Antarctic ice cores at the Law Dome (Etheridge et al., 1998),
instrumental observations at Mauna Loa (Keeling et al., 2008), and p(CO2) concentrations resulting from SRES
climate scenarios using the Bern carbon cycle model (Nakićenović & Swart, 2000).
outlined scenarios projecting fossil fuel CO2 emissions through 2100, based on population and economic
growth, rates of technology development, and societal attitudes. SRES scenarios indicate a continual rise
in p(CO2)atm with levels at 2100 ranging between ~530 to 970 ppm p(CO2)atm (Figure 3.3), depending on the
scenario and carbon cycle model used. Recent models indicate that continued warming of the sea surface will
inhibit ocean carbon uptake, leading to ~4% higher atmospheric carbon dioxide levels in 2100 than expected
by the SRES scenarios (Plattner et al., 2001). Atmospheric CO2 trajectories beyond 2100 may exceed 1000
ppm, or could approach stabilisation targets from 450 to as high as 1000 ppm, roughly along WRE stabilisation
pathways (Wigley et al., 1996).
Environmental variability during the recent evolutionary history of marine ecosystems is also important to
consider for ocean acidification studies, due to its role in shaping the physiological tolerances of marine
organisms and maintaining genetic diversity within populations. Atmospheric CO2 has varied greatly over Earth
history (Kasting, 1993), and generally decreased through the Phanerozoic from over 5000 ppm to relatively
stable and low levels well below 1000 ppm through most of the Cenozoic (Berner, 1990). Through at least
the latter half of the Quaternary (~0.8 My), atmospheric CO2 has generally oscillated rhythmically between
glacial (~180 ppm) to interglacial (~280 ppm) extremes (Figure 3.4). This pendulum of p(CO2)atm has driven
a parallel modulation of ocean pH, temperature, hypoxia, and other factors, that undoubtedly influenced the
recent evolution of a many marine organism. Variation in ocean chemistry during this period is therefore an
important context for considering the impacts of future climate scenarios. Although few experimental studies
have included p(CO2)atm levels below ambient, Riebesell et al. (2000) included a “glacial” climate (190 ppm
CO2) along with present-day (350 ppm) and future (750 ppm) levels in a mescosm study concerning their
effects on phytoplankton communities.
57
Part 2: Experimental design of perturbation experiments
Figure 3.4 Quaternary atmospheric carbon dioxide (a) based on ice core records (Petit et al. 2001) and estimated
ocean pH (b) (J. Barry, unpubl.).
Levels of p(CO2)atm exceeding those expected under future climate scenarios may also be useful, for example in
studies examining the physiological response of organisms to environmental hypercapnia. Variation in ocean p(CO2),
pH and other key parameters can act directly on the physiological performance of organisms, with impacts on the
growth, reproduction, behaviour and survival of individuals, which in turn affect the demographic rates (i.e. birth and
death rates) of populations, interactions between species and, ultimately, the structure and function of ecosystems.
While the objectives of an ocean acidification research project include characterisation of the performance of
organisms over a realistic range of future p(CO2)atm levels, inclusion of considerably higher p(CO2) treatments often
help constrain their boundaries of performance (e.g. Kurihara & Shirayama, 2004). Knowledge of performance
boundaries can also guide subsequent studies on similar taxa. This approach may be particularly important for
studies where logistical constraints limit replication or variation among replicates reduces statistical power.
Key p(CO2)atm values could be organised according to arbitrary schemes, such as multiples of preindustrial
atmospheric levels (PAL) or a range of near log-ratio values (Table 3.2). Multiples of PAL allow 3 or 4 treatment
levels in the realm of realistic changes in p(CO2)atm over the next millennium, and relate more closely to the
much larger changes in p(CO2)atm levels that have occurred through Earth history (e.g. Kasting, 1993). Other
schemes, such as a ~log-ratio method shown in Table 3.2 could span even larger ranges, but provide even fewer
treatment levels within reasonable future climate conditions. The range of key values using the latter examples
overlaps only marginally with the range of probable future p(CO2)atm levels, reducing their relevance for the
general goals of ocean acidification research. Refer to chapter 4 for further information on statistics.
58
Table 3.2 Alternative designs for key atmospheric p(CO2) values used in ocean acidification studies.
• Realistic range of atmospheric p(CO2) levels including glacial, preindustrial (280 ppm) and
intermediate values to 2100 and beyond (1000 ppm; Wigley, 1996; Nakićenović & Swart 2000):
180, 280, 380, 450, 550, 650, 750, 1000
• Integer multiples of 280 ppm (pre-industrial value):
280, 560, 840, 1120, 1400, 1680, 1960, 2240, 2520, 2800, 4480, …
• Range of ~log-ratio values to investigate the response to exposure to very high p(CO2)atm levels:
80, 1000, 3000, 10000, 30000, 100000, ...
3.2.2
Conversion of atmospheric p(CO2) levels to corresponding in situ ocean chemistry
Key atmospheric p(CO2) values can be defined and used as guidelines, but their corresponding values for ocean
carbonate system parameters are the primary measurements for ocean acidification experiments, and should also be
reported. How can investigators convert key atmospheric p(CO2) values to the in situ p(CO2), pH, Ωa, Ωc, or other
carbonate system parameters of interest for specific ocean acidification experiments? The simplest method may
be to use published predictions of future ocean carbonate system values, based on changes in atmospheric p(CO2)
(Gattuso & Lavigne, 2009). Examples include predictions of future changes in ocean pH and carbonate saturation
states (Gattuso et al., 1999, Caldeira & Wickett, 2003, 2005; Guinotte et al., 2003; Feely et al., 2004; Orr et al.,
2005; Kleypas et al., 2006; Hoegh-Guldberg et al., 2007). It is worth noting that the present-day latitudinal trends
in some carbonate system variables will exist in the future, as seen for example in Figure 3.5 (Orr et al., 2005). If
all recoverable fossil fuels are eventually burnt, pH is expected to decline by about 0.8 units from the pre-industrial
total, leading to an average surface pH decline from about 8.2 down to 7.4 (Caldeira & Wickett, 2003).
An alternate approach for predicting future ocean carbonate system values is to assume that surface water is in
equilibrium with the atmosphere and use software such as co2sys (http://cdiac.ornl.gov/oceans/co2rprt.html),
seacarb (http://www.obs-vlfr.fr/~gattuso/seacarb.phphttp://cran.at.r-project.org/web/packages/seacarb/), or
csys (http://www.soest.hawaii.edu/oceanography/faculty/zeebe_files/CO2_System_in_Seawater/csys.html).
Each of these software packages can calculate a series of ocean carbonate parameters based on input for other
carbonate system factors (see chapter 2 for examples of calculations using seacarb). Knowledge of 2 of 4 key
carbonate parameters (p(CO2), pH, DIC, AT) and a few physical factors (temperature, salinity, [SiO2], [ PO 3−
4 ], depth)
is sufficient to calculate all other carbonate system values (see Dickson, this volume, for detailed information). This
approach is particularly important for locations differing from the open ocean, such as coastal zones, inland seas,
oxygen minimum zones, and deep-sea environments. Large divergence of the concentrations
of calcium [Ca2+] and/
€
or magnesium [Mg2+] from normal seawater values affects the calculation of calcite saturation state from [CO 2−
3 ]
(Tyrrell & Zeebe, 2004). For some habitats and microenvironments, p(CO2)atm can be considerably higher than
observed atmospheric levels.
€
Using carbonate chemistry software, investigators can substitute a future atmospheric carbon dioxide
level (e.g.
750 ppm) for p(CO2) in surface waters, then combine this with AT to approximate future DIC, Ωa, Ωc, or pH.
Because p(CO2) varies over the world ocean (Figure 3.1), a somewhat more accurate use of this method would
use the p(CO2) in surface waters at the location of interest, then increase it by an increment corresponding
to the increase in p(CO2)atm of interest (e.g. +365 ppm: the difference between 750 and 385 ppm). Carbonate
system parameters could then be calculated assuming no change in AT. For deeper waters, the increase in DIC
calculated for the surface could be added to observed deep-water DIC levels, and combined with in situ AT to
calculate other carbon system elements. For example, 30°C surface water with a total alkalinity of 2264 µmol
kg-1 equilibrated with a 380 ppm atmosphere, has a DIC concentration of ~1919 µmol kg-1 and an Ωa of 3.94,
compared to a similar sample at 5°C, in which DIC is 2107 µmol kg-1 and Ωa drops to only 1.74. Conversion
59
Part 2: Experimental design of perturbation experiments
of atmospheric targets to deep-sea values can
produce even larger changes. Due to low water
temperatures and the accumulation of respiratory
CO2 at depth, p(CO2)atm values of 380 and 750
ppm can be equivalent to bathyal p(CO2) values of
1000 and >3000 µatm, respectively, particularly
in oxygen minimum zones. Atmospheric p(CO2)
is typically reported for dry air, and should be
adjusted for the vapour pressure of water (100%
humidity) near the ocean surface (Zeebe & WolfGladrow, 2001), leading to a decrease in p(CO2)
of about 3%.
Factors that will affect the predicted carbonate
system parameters at key p(CO2)atm values include
location, temperature, depth, productivity, regional
to local oceanographic dynamics, the stability
of the climate system and its recent history. The
heterogeneous distribution of carbonate chemistry
in the world ocean derives from the dynamic quasiequilibrium that exists currently in the atmosphereocean system. As indicated before, even though
p(CO2)atm is relatively homogenous over the globe,
p(CO2) in surface waters is highly heterogeneous
(Figure 3.1). Estimating accurately the change in
p(CO2) at a particular location under an atmospheric
Figure 3.5 Predicted changes in future atmospheric
level of 750 ppm CO2, particularly in waters deep
p(CO2), ocean pH and ocean carbonate ion concentrations.
beneath the surface, is not straightforward. The
‘I’ in b and c refers to the IS92a future scenario, ‘S’ to
predicted values, particularly for deep-sea waters
the S650 scenario. (From Orr et al. (2005). Reprinted
by permission from MacMillan Publishers Ltd: Nature,
that respond slowly to recent changes in atmospheric
Copyright (2005)).
CO2, depend upon the stability of the climate and
the time scales for equilibration for the atmosphereocean carbonate system. Failures of other assumptions, such as the long-term stability of total alkalinity (e.g. Ilyina
et al., 2009), ocean ventilation, carbonate rain ratios, etc. under continued climate change, also affect the accuracy
of carbonate system predictions.
3.2.3
The number of treatments in ocean acidification experiments
The ideal perturbation experiment would measure the response of the experimental system (e.g. phytoplankton
community dynamics, coral calcification or animal physiology) to a range of pH (or other carbonate system
parameters) corresponding to a series of atmospheric p(CO2) from 180 to 1000 ppm or higher. This would provide
an understanding of the performance of organisms under glacial climate, present-day, and the near future, including
the trend over a range in p(CO2)atm as well as inflection points in performance that could indicate tipping points.
In practice, however, logistical considerations limit the number of treatment levels and replication, thereby
reducing the predictive resolution of the results. How many treatments levels are sufficient? What statistical
design is optimal? At a minimum, 2 or 3 treatments levels analysed using an analysis of variance (ANOVA)
design can yield valuable comparisons among locations or pH treatments (e.g. Manzello et al., 2008). Analysis
of variance is often used to compare differences among categorical variables, and may be the optimal design
for studies limited to a few treatments. However, when the number of treatment levels can be increased,
regression designs are typically more advantageous, particularly for continuous variables (e.g. ocean pH).
60
Since the parameters and processes measured are continuous variables, regression designs, particularly those
with many treatment levels, allow interpretations of results that define functional relationships (e.g. variation
in animal performance over a range of p(CO2) levels), and may be more effective in identifying tipping points,
if they exist. Although three treatments is a minimum for regression analyses, additional treatments, even
at the expense of replication within treatments, often provides greater inferential power for a continuous
variable. This approach may be particularly valuable in detecting potential tipping points. Regression may be
less effective for systems where within-treatment responses are variable, and analysis of variance designs may
increase statistical power. A more thorough discussion of experimental design is presented in chapter 4.
Where to start? At a minimum, perturbation experiments typically compare one or more treatments simulating
future atmospheric CO2 levels to a baseline control treatment. A “preindustrial” climate near 280 ppm p(CO2)atm
is the most suitable baseline treatment, since it represents a long-term (i.e. millennial) average concentration that
has shaped animal performance and ecosystem function. Ambient or “present-day” p(CO2)atm (385 ppm CO2 =
2008 average on Mauna Loa; http://cdiac.ornl.gov/ftp/trends/co2/maunaloa.co2) has also been used as a baseline
treatment in various studies, for comparing animal performance with future, higher p(CO2)atm levels. However,
because the amount of anthropogenic carbon dioxide in the atmosphere has doubled nearly every 31 years (Hofmann
et al., 2008), “present-day” is a rapidly shifting target, and has already exceeded a p(CO2)atm threshold (350 ppm)
proposed as a potential ecological tipping point (Hansen et al., 2008).
Guidelines presented here (Table 3.3) are based on the number of treatment levels that can be supported both
technically and financially. For studies limited to very few perturbed treatment levels, comparisons between “presentday” values (currently ~385 ppm), and 750 ppm (“future”) are recommended as primary treatment levels. While
“preindustrial” (280 ppm) p(CO2)atm may be more relevant as a control treatment for many organisms than presentday p(CO2)atm, 280 ppm is technically difficult to achieve for most experiments, and “present-day” can be substituted
as an unperturbed, control treatment. Ambient p(CO2)atm has value as a p(CO2)atm treatment because natural systems
have acclimated to this level over decadal time scales, and because it provides a context for current changes in
response variables. For a future treatment, 750 ppm is favoured after considering that ongoing efforts to curb fossil
fuel CO2 emissions, recent emission records, and climate modelling based on (Nakićenović & Swart, 2000; Plattner
et al., 2001) indicate p(CO2)atm will reach or exceed 750 ppm by 2100, a value midway between SRES scenario A1B
and A2 (Figure 3.3). In addition, a 2-treatment comparison between these values is more likely to detect significant
changes in performance than between smaller p(CO2)atm changes, particularly for field experiments where withintreatment variability can be high. Inclusion of “preindustrial” p(CO2)atm (280 ppm) conditions should be considered
as a third treatment. Using this 2 or 3-treatment design, it should be possible to document changes to date (i.e. since
~1850), and project changes likely to occur by the end of this century.
As the number of treatment levels increases, the addition of stabilisation targets and important potential tipping
points (e.g. 350 ppm: Hansen et al., 2008; 450 ppm: Hoegh-Guldberg et al., 2007, McNeil & Matear, 2008) for
atmospheric CO2 can be added. Tipping points can be crucial for climate policy development and for increased
awareness of society to the potentially non-linear response of Earth systems to climate change (Lenton et al., 2008).
Key values of atmospheric p(CO2) recommended for ocean acidification studies (Table 3.3) can be used as guidelines
for the design of experiments. These guidelines may require modification to fit particular needs, such as the addition
of higher values to examine the boundaries of animal performance or to allow closer correspondence with crucial
levels for specific carbonate system parameters (e.g. Ωa or Ωc ~1). Increasing the number of treatments will provide
greater predictive power, particularly for non-linear responses to ocean acidification.
Even if this chapter is devoted to a discussion on atmospheric CO2 targets, it is nevertheless important to
note the potential synergy between multiple stressors. The effects of ocean acidification, thermal stress, and
expanding hypoxia, all linked to anthropogenic climate change, act together to constrain the window of
performance for marine organisms (Pörtner, 2008). Therefore, consideration of multiple climate stressors for
ocean acidification experiments will elevate the value of these studies, allowing an integrated view of climate
change impacts on ocean ecosystems.
61
Part 2: Experimental design of perturbation experiments
Table 3.3 Key p(CO2)atm values (ppm) for ocean acidification studies. These (CO2)atm levels are useful guidelines
for perturbation experiments, and can be supplemented with other values of importance for specific studies, such
as higher values for evaluating animal performance, or adjustments to correspond to key carbonate system values
(e.g. Ωa or Ωc ~1).
# of Treatments
3.3
2
present-day (~385), 750
3
280, present-day, 750
4
280, present-day, 550, 750
6
280, present-day, 550, 650, 750, 1000
8
180, 280, present-day, 450, 550, 650, 750, 1000
>8
Add values (e.g. 350, other) to increase resolution
Strengths and weaknesses
−
Designing ocean acidification experiments in the context of realistic ranges in future atmospheric
carbon dioxide levels will increase our understanding of the effects of impending environmental
change and enable information-based policy development for climate adaptation.
−
Use of key atmospheric carbon dioxide values as the principal treatment levels for ocean acidification
experiments will provide a strong link between ocean acidification science and climate policy
discussion.
−
Comparison of results among ocean acidification studies will be easier by using common atmospheric
CO2 targets, even though ocean carbonate chemistry parameters may differ.
−
Regression designs with higher numbers of ocean acidification treatment levels are, in general, more
likely to identify tipping points in organism or community performance than those with fewer treatment
levels.
3.4
62
Recommended p(CO2)atm levels
Potential pitfalls
−
Conversion of key atmospheric p(CO2) values to specific parameters of the ocean carbonate system
can be complex.
−
Characterisation of the effects of tipping points will be difficult, even if speculations about such values
exist. Tipping points (p(CO2)atm) likely differ from single organisms to whole ecosystems with complex
direct and indirect consequences, time scales, and effects. On the other hand, some studies indicate a
tipping point of 450 ppm for coral reefs, and if this point is not included, important results might be
neglected.
−
Using too narrow a scope of p(CO2)atm levels under conditions of high within-treatment variability may
reduce the statistical power of analyses of variance.
−
Physiological processes depend strongly on temperature and are often subject to a narrow thermal
window (Pörtner et al., 2004). This should not be neglected and multifactor experiments studying the
interacting effects of p(CO2)atm, temperature, and perhaps oxygen should, if possible, be conducted.
See chapter 9 of this Guide for more information.
−
Logistical and other constraints limiting the number of p(CO2) treatments can reduce inference
concerning non-linear functional responses of organisms to ocean acidification.
3.5
Suggestions for improvements
Society and the field of ocean acidification science will benefit from the standardisation of atmospheric carbon
dioxide levels as the common currency for discussion of perturbation studies. Use of atmospheric levels
will promote the effective communication of results from ocean acidification studies to policymakers, and
increase the impact of ocean acidification science in the development of climate adaptation policy. Within
the oceanographic community the combined use of atmospheric p(CO2) levels and related ocean carbonate
chemistry parameters will help standardise comparisons of the potential effects of ocean acidification among
habitats and ecosystems.
Increased focus on the characterisation of functional responses through the use of higher numbers of treatment
levels will also increase the relevance of ocean acidification studies for society. Higher resolution among
treatment levels will help characterise non-linear relationships between p(CO2)atm levels and ecosystem
processes, and help evaluate ecosystem function near proposed tipping points.
3.6
Data reporting
The primary goal of data reporting for climate targets is to provide a template for comparing experimental
results within the ocean science community, as well as between ocean and terrestrial systems. The
common currency of atmospheric carbon dioxide levels and the use of standard or key p(CO2)atm values
for most studies will elevate the value of ocean acidification science for society. Towards that goal,
ocean acidification studies should report carefully the p(CO2)atm levels of interest for the study, with the
corresponding levels of the various parameters of the carbonate system. Atmospheric carbon dioxide
values should be used in the abstract, discussion, and conclusions of papers to maximise the likelihood
that terms common within climate discussions are used clearly. Also required is a careful reporting of
relevant ocean carbonate chemistry. While some values (p(CO2)) may be similar to target values, others
(e.g. carbonate saturation, bicarbonate levels, etc.) may not. Considering the growing concern for the
combined impacts of warming, acidification, and environmental hypoxia, it is advisable to report these
and other potentially relevant environmental parameters.
3.7
Recommendations for standards and guidelines
1. Ocean acidification experiments should be based primarily on a broad range of realistic p(CO2)atm
values spanning glacial, present-day, and the future (1000+ ppm).
2. Atmospheric p(CO2) values of 180, 280, 350, “present-day”, 450, 550, 650, 750, and 1000 are
guidelines for designing ocean acidification experiments.
3. For logistically limited studies, primary targets of 280, present-day (currently 385), and 750 ppm
should be used.
4. Values of p(CO2)atm exceeding realistic ranges can be useful to examine the boundaries of animal
performance, but should be followed by or coupled with realistic values.
5. Key atmospheric carbon dioxide levels should be converted to corresponding values of in situ
ocean carbonate parameters for specific ocean acidification experiments.
6. Key p(CO2)atm and in situ carbonate system variables should be reported in tandem in publications
reporting the results of ocean acidification experiments.
7. An increased number of treatments, even with reduced replication, may have greater power to
characterise the functional relationship between ocean acidification parameters and organism or
ecosystem performance.
8. Use of increased treatment levels increases the likelihood of detecting “tipping points” in organism
or community performance.
63
Part 2: Experimental design of perturbation experiments
3.8
References
Antonov J. I., Locarnini R. A., Boyer T. P., Mishonov A. V. & Garcia H. E., 2006. World Ocean Atlas 2005.
Volume 2: Salinity. In: Levitus S. (Ed.), NOAA Atlas NESDIS 62, pp. 1-182. Washington, D.C.: U.S.
Government Printing Office.
Bensoussan N. & Gattuso J.-P., 2007. Community primary production and calcification in a NW Mediterranean
ecosystem dominated by calcareous macroalgae. Marine Ecology Progress Series 334:37-45.
Berner R. A., 1990. Atmospheric carbon dioxide levels over Phanerozoic time. Science 249:1382-1386.
Boyer T. P., Stephens C., Antonov J. I., Conkright M. E., Locarnini R. A., O’Brien T. D. & Garcia H. E., 2002.
World Ocean Atlas 2001, Volume 2: Salinity. In: Levitus S. (Ed.), NOAA Atlas NESDIS 50, pp. 1-165.
Washington D.C.: U.S. Government Printing Office.
Buitenhuis E. T., de Baar H. J. W. & Veldhuis M. J. W., 1999. Photosynthesis and calcification by Emiliania huxleyi
(Prymnesiophyceae) as a function of inorganic carbon species. Journal of Phycology 35:949–959.
Caldeira K. & Wickett M., 2003. Anthropogenic carbon and ocean pH. Nature 425:365.
Caldeira K. & Wickett M. E., 2005. Ocean model predictions of chemistry changes from carbon dioxide
emissions to the atmosphere and ocean. Journal of Geophysical Research- Oceans 110, C09S04.
doi:10.1029/2004JC002671.
Etheridge D. M., Steele L. P., Langenfelds R. L., Francey R. J., Barnola J.-M., & Morgan V. I., 1998. Historical
CO2 records from the Law Dome DE08, DE08-2, and DSS ice cores. In Trends: A compendium of data
on global change. Oak Ridge, Tenn., U.S.A.: U.S. Department of Energy.
Feely R. A., Sabine C. L., Hernandez-Ayon J. M., Ianson D. & Hales B., 2008. Evidence for upwelling of
corrosive “acidified” water onto the continental shelf. Science 320:1490-1492.
Feely R. A., Sabine C. L., Lee K., Berelson W., Kleypas J., Fabry V. J. & Millero F. J., 2004. Impact of
anthropogenic CO2 on the CaCO3 system in the oceans. Science 305:362-366.
Findlay H. S., Tyrrell T., Bellerby R. G. J., Merico A. & Skjelvan I., 2008. Carbon and nutrient mixed layer
dynamics in the Norwegian Sea. Biogeosciences 5:1395–1410.
Friederich G. E., Walz P. M., Burczynski M. G. & Chavez F. P., 2002. Inorganic carbon in the central California
upwelling system during the 1997-1999 El Niño-La Niña event. Progress in Oceanography 54:185-203.
Garcia H. E., Locarnini R. A., Boyer T. P. & Antonov J. I., 2006. World Ocean Atlas 2005. Volume 4: Nutrients
(phosphate, nitrate, silicate). In: Levitus S. (Ed.), NOAA Atlas NESDIS 64, pp. 1-396. Washington, D.C.:
U.S. Government Printing Office.
Gattuso J.-P., Allemand D. & Frankignoulle M., 1999. Photosynthesis and calcification at cellular, organismal
and community levels in coral reefs: a review on interactions and control by carbonate chemistry.
American Zoologist 39:160-183.
Gattuso J.-P. & Lavigne H., 2009. Technical Note: Approaches and software tools to investigate the impact of
ocean acidification. Biogeosciences 6:2121-2133.
Guinotte J. M., Buddemeier R. W. & Kleypas J. A., 2003. Future coral reef habitat marginality: temporal and
spatial effects of climate change in the Pacific basin. Coral Reefs 22:551-558.
Hansen J., Sato M., Kharecha P., Beerling D., Berner R., Masson-Delmotte V., Pagani M., Raymo M., Royer D.
L. & Zachos J. C., 2008. Target atmospheric CO2: Where should humanity aim? The Open Atmospheric
Science Journal 2:217-231.
Hoegh-Guldberg O., Mumby P. J., Hooten A. J., Steneck R. S., Greenfield P., Gomez E., Harvell C. D., Sale P.
F., Edwards A. J., Caldeira K., Knowlton N., Eakin C. M., Iglesias-Prieto R., Muthiga N., Bradbury R.
H., Dubi A. & Hatziolos M. E., 2007. Coral reefs under rapid climate change and ocean acidification.
Science 318:1737-1742.
Hofmann D. J., Butler J. H. & Tans P. P., 2008. A new look at atmospheric carbon dioxide. Atmospheric
Environment 43:2084-2086.
Ilyina T., Zeebe R. E., Maier-Reimer E. & Heinze C., 2009. Early detection of ocean acidification effects on
marine calcification. Global Biogeochemical Cycles 23, GB1008. doi:10.1029/2008GB003278.
64
Kasting J., 1993. Earth’s early atmosphere. Science 259:920-926.
Keeling R. F., Piper S. C., Bollenbacher A. F. & Walker J. S., 2008. Atmospheric CO2 records from sites in the
SIO air sampling network. In Trends: A compendium of data on global change. Oak Ridge, Tenn.: Oak
Ridge National Laboratory, U.S. Department of Energy.
Key R. M., Kozyr A., Sabine C. L., Lee K., Wanninkhof R., Bullister J. L., Feely R. A., Millero F. J., Mordy
C. & Peng T. H., 2004. A global ocean carbon climatology: results from Global Data Analysis Project
(GLODAP). Global Biogeochemical Cycles 18, GB4031. doi:10.1029/2004GB002247.
Kikkawa T., Ishimatsu A. & Kita J., 2003. Acute CO2 tolerance during the early developmental stages of four
marine teleosts. Environmental Toxicology 18:375-382.
Kleypas J. A., Feely R. A., Fabry V. J., Langdon C., Sabine C. L. & Robbins L. L., 2006. Impacts of ocean
acidification on coral reefs and other marine calcifiers: a guide for future research. 88 p. Boulder,
Colorado: Institute for the Study of Society and Environment (ISSE) of the University Corporation for
Atmospheric Research (UCAR).
Kurihara H. & Shirayama Y., 2004. Effects of increased atmospheric CO2 on sea urchin early development.
Marine Ecology Progress Series 274:161-169.
Lenton T. M., Held H., Kriegler E., Hall J. W., Lucht W., Rahmstorf S. & Schellnhuber H. J., 2008.
Tipping elements in the Earth’s climate system. Proceedings of the National Academy of Science
U.S.A. 105:1786-1793.
Locarnini R. A., Mishonov A. V., Antonov J. I., Boyer T. P. & Garcia H. E., 2006. World Ocean Atlas 2005.
Volume 1: Temperature. In: Levitus S. (Ed.), NOAA Atlas NESDIS 61, pp. 1-182. Washington, D.C.:
U.S. Government Printing Office.
Manzello D. P., Kleypas J. A., Budd D. A., Eakin C. M., Glynn P. W. & Langdon C., 2008. Poorly cemented
coral reefs of the eastern tropical Pacific: Possible insights into reef development in a high-CO2 world.
Proceedings of the National Academy of Science U.S.A. 105:10450-10455.
McNeil B. I. & Matear R., 2008. Southern Ocean acidification: a tipping point at 450-ppm atmospheric CO2.
Proceedings of the National Academy of Science U.S.A. 105:18860-18864.
Nakićenović N. & Swart R., 2000. Special report on emissions scenarios: a special report of Working Group III
of the Intergovernmental Panel on Climate Change, 599 p. Cambridge: Cambridge University Press.
Orr J. C., Fabry V. J., Aumont O., Bopp L., Doney S. C., Feely R. A., Gnanadesikan A., Gruber N., Ishida A.,
Joos F., Key R. M., Lindsay K., Maier-Reimer E., Matear R., Monfray P., Mouchet A., Najjar R. G.,
Plattner G. K., Rodgers K. B., Sabine C. L., Sarmiento J. L., Schlitzer R., Slater R. D., Totterdell I. J.,
Weirig M. F., Yamanaka Y. & Yool A., 2005. Anthropogenic ocean acidification over the twenty-first
century and its impact on calcifying organisms. Nature 437:681-686.
Petit J. R., Jouzel J., Raynaud D., Barkov N. I., Barnola J.-M., Basile I., Bender M., Chappellaz J., Davis
J., Delaygue G., Delmotte M., Kotlyakov V. M., Legrand M., Lipenkov V., Lorius C., Pépin L., Ritz
C., Saltzman E. & Stievenard M., 2001. Vostok ice core data for 420,000 years. IGBP PAGES/World
Data Center for Paleoclimatology Data Contribution Series #2001-076: Boulder, CO: NOAA/NGDC
Paleoclimatology Program.
Plattner G. K., Joos F., Stocker T. F. & Marchal O., 2001. Feedback mechanisms and sensitivities of ocean
carbon uptake under global warming. Tellus B 53:564-592.
Pörtner H.-O., 2008. Ecosystem effects of ocean acidification in times of ocean warming: a physiologist’s
view. Marine Ecology Progress Series 373:203-217.
Pörtner H.-O., Langenbuch M. & Reipschlager A., 2004. Biological impact of elevated ocean CO2 concentrations:
Lessons from animal physiology and earth history. Journal of Oceanography 60:705-718.
Riebesell U., Zondervan I., Rost B., Tortell P. D. & Morel F. M. M., 2000. Reduced calcification of marine
plankton in response to increased atmospheric CO2. Nature 407:364-367.
Stephens, C., J. I. Antonov, T. P. Boyer, M. E. Conkright, R. A. Locarnini, T. D. O’Brien, H. E. Garcia, 2002.
World Ocean Atlas 2001, Volume 1: Temperature. In: Levitus S. (Ed.), NOAA Atlas NESDIS 49, pp.
1-167. Washington D.C.: U.S. Government Printing Office.
65
Takahashi T., Sutherland S. C., Wanninkhof R., Sweeney C., Feely R. A., Chipman D. W., Hales B., Friederich
G., Chavez F., Sabine C., Watson A., Bakker D. C. E., Schuster U., Metzl N., Inoue H. Y., Ishii M.,
Midorikawa T., Nojiri Y., Körtzinger A., Steinhoff T., Hoppema M., Olafsson J., Arnarson T. S., Tilbrook
B., Johannessen T., Olsen A., Bellerby R., Wong C. S., Delille B., Bates N. R. & de Baar H. J. W., 2009.
Climatological mean and decadal change in surface ocean pCO2, and net sea-air CO2 flux over the global
oceans. Deep-Sea Research II 56:554-577.
Tyrrell T. & Zeebe R. E., 2004. History of carbonate ion concentration over the last 100 million years.
Geochimica et Cosmochimica Acta 68:3521-3530.
Wigley T. M. L., Richels R. & Edmonds J. A., 1996. Economic and environmental choices in the stabilization
of atmospheric CO2 concentrations. Nature 379:240-243.
Wooton J. T., Pfister C. A. & Forester J. D., 2008. Dynamic patterns and ecological impacts of declining ocean
pH in a high-resolution multi-year dataset. Proceedings of the National Academy of Science U.S.A.
105:18848-18853.
Zeebe R. E. & Wolf-Gladrow D. A., 2001. CO2 in seawater: equilibrium, kinetics, isotopes. 346 p. Amsterdam:
Elsevier.
66